A Histochemical Approach to Correlative Light and Electron Microscopic Detection of Acidic Glycoconjugates by a Sensitized High Iron Diamine Method

Abstract

A sensitized high iron diamine method is among the reliable and useful histochemical means of detecting acidic glycoconjugates by light microscopy. Because the final reaction products obtained using this method are heavy metals, it can be applied to specimens for visualization by both light and electron microscopy. In this study the high iron diamine method was utilized successfully as a correlative light and electron microscopic method for detection of acidic glycoconjugates.

Keywords

histochemistry correlative light and electron microscopy acidic glycoconjugates sensitized high iron diamine

To fully elucidate the structural and functional details of cells and tissues, it is often necessary to extend results obtained by light microscopy to the ultrastructural level. Because the procedures for tissue preparation and staining are usually quite different, it is rather difficult to examine the same specimen by both light and electron microscopy. Since 1981, several methods have been developed to correlate light and electron microscopic examinations histologically and histochemically (Kushida and Kushida 1981; Nagato et al. 1984; Nagato and Kushida 1985; Rieder and Bowser 1983, 1985; Kushida et al. 1993). However, because suitable methods common to both types of microscopy have not yet been developed for identifying acidic glycoconjugates, only a few attempts to detect such carbohydrates in the same histological site have been reported (Nagato et al. 1984; Nagato and Kushida 1985). The sensitized high iron diamine (S-HID) method was developed to detect small amounts of acidic glycoconjugates in tissues by light microscopy (Hirabayashi 1992). Because the final reaction products of the S-HID method are heavy metals, it appears plausible that this method could be adapted for light and electron microscopic histochemistry of acidic glycoconjugates in the same specimen. Here we report the development of a correlated light and electron microscopic approach using the S-HID method for detecting acidic glycoconjugates in the same tissues.

Materials and Methods

Tissue Preparation for Light Microscopy

After ether anesthesia, six male Wistar rats were perfused via the left ventricle with a physiological saline solution containing 2.5 mg/dl heparin, followed by perfusion with 4.0% paraformaldehyde/0.1% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) containing 7.5% sucrose (fixative solution) for 10 min at room temperature (RT). Kidneys were dissected out, cut into tiny pieces (7 X 7 X 3 mm), and immersed in the same fixative solution overnight (for approximately 16 hr) at 4C. After rinsing with 0.1 M phosphate buffer (pH 7.4) containing 7.5% sucrose, the tissues were dehydrated in an ethanol series of ascending concentrations, immersed in methyl benzoate, cleared in xylene, and embedded in paraffin. Sections were cut at a thickness of approximately 4 μm, mounted on glass slides coated with silane (3-aminopropyl triethoxysilane), and dried in an oven for 5–6 hr at 60C.

Staining Procedure

A sensitized high iron diamine (S-HID) method was employed. Detailed procedures for the S-HID method have been described elsewhere (Hirabayashi 1992). Briefly, deparaffinized and rehydrated sections were incubated in an HID solution for 60 min at 30C, immersed in 0.5 mM potassium trichloro(ethylene)platinum solution for 60 min at RT, reduced within 0.2% sodium borohydride solution for 30 sec at RT, subjected to a silver enhancement procedure for 8 min at 20C, and then incubated in an appropriate photographic fixative for 2 min. Stained sections were rinsed thoroughly with tapwater, dehydrated in a graded ethanol series, immersed in QY-1 (N-butyl glycidyl ether) (Nisshin EM; Tokyo, Japan), and mounted in an Epon 812 mixture with a coverslip. The sections were examined and photographed using a light microscope (PROVIS AX 70 with U-PHOTO; Olympus, Tokyo, Japan).

Enzyme Digestion

Before the S-HID staining, some tissue sections were subjected to digestion with heparitinase to identify heparan sulfate. Hydrated sections were incubated in 0.1 U/ml of heparitinase in 0.1 M phosphate buffer (pH 7.0) for 16 hr at 45C (Hovingh and Linker 1970). As a control for the enzyme digestion, sections were exposed to 0.1 M phosphate buffer without enzyme.

Tissue Preparation for Electron Microscopy

After light microscopic examination, S-HID-stained sections were exposed by removing the coverslips in QY-1 and wiping up excessive QY-1 surrounding the tissue section. An Epon mixture was dropped on the tissue sections, and eventually Epon 812 mixture-filled beem capsules were placed directly on the exposed sections and polymerized in an oven for 24 hr at 60C. The resin blocks containing the stained sections were removed from the glass slides by heating at 60C. Ultrathin sections were cut at a thickness of approximately 80-100 nm on an ultramicrotome (Ultracut E; C. Reicherlt-Jung Optische Werke, Wien, Austria) and mounted on copper grids. The sections, with or without counterstaining with uranyl acetate and lead salts, were examined in a transmission electron microscope (JEM 1200 EX; JEOL Tokyo, Japan).

Figure 1

When a rat kidney is stained with S-HID, renal glomeruli, Bowman's capsules, and basement membranes of urinary tubule epithelial cells react strongly (light micrograph). Bar = 50 μm.

Figure 2

Identical site of rat renal glomeruli to that in Figure 1. S-HID-stained ultrathin sections are counterstained with uranyl acetate and lead salts (electron micrograph). Bar = 50 μm.

Figure 3

A high magnification of the rectangle-outlined area in Figure 2. The basement membranes of Bowman's capsule (arrow) and glomerulus (arrowheads) are stained vividly with S-HID (electron micrograph). Bar = 1 μm.

Figure 4

Part of a renal glomerular basement membrane in a section from Figure 2. A large number of silver grains are scattered on the basement membrane (arrows) and surface coat of podocytes and endothelial cells (electron micrograph). Bar = 1 μm.

Results

By light microscopy, a number of histological structures exhibited strong positive reactions with S-HID, including the glomerular basement membrane, mesangial matrices and the surface coat of podocytes and vascular endothelial cells, basement membranes of Bowman's capsules and urinary tubules, and basal infoldings of urinary tubule epithelial cells (Figure 1). Examination of the same sections by electron microscopy revealed satisfactory preservation of histological structures (Figure 2). At lower magnification (Figure 2), the basement membranes of urinary tubules showed positive S-HID reaction. At higher magnifications (Figures 3 and 4), S-HID-reactive histological structures in the glomeruli were recognized to be basement membranes of renal glomeruli and Bowman's capsules, mesangial matrices, certain cytoplasmic granules, and the surface coat of podocytes and vascular endothelial cells.

When visualized by light microscopy, heparitinase digestion presented a variable decrease in S-HID staining intensity in the basement membranes of renal glomeruli, Bowman's capsule, and urinary tubules (Figure 5). However, positive S-HID reactions in the glomerular mesangial matrices were decreased only slightly in intensity (Figure 5). At the ultrastructural level (Figures 6 and 7), the S-HID reaction in the basement membranes of the same renal glomeruli was either abolished or markedly diminished in intensity. However, large amounts of S-HID reaction product persisted in the cytoplasm and surface coat of podocytes and vascular endothelial cells after digestion with heparitinase (Figures 6 and 7). In control tissue sections for the digestion with heparitinase, images obtained by the S-HID staining were almost identical to those in intact sections (not shown).

Figure 5

When a rat kidney is stained with S-HID after digestion with heparitinase, positive reactions in renal glomeruli, Bowman's capsules, and basal laminae of urinary tubule epithelial cells are variably decreased in intensity (light micrograph). Bar = 50 μm.

Figure 6

A high magnification of the boxed area in Figure 5. The ultrathin sections are counterstained with uranyl acetate and lead salts. The basement membranes of glomeruli (arrowheads) show greatly reduced reactivity with S-HID (electron micrograph). Bar = 1 μm.

Figure 7

Part of the renal glomerular basement membrane in a section from Figure 6. Silver grains are lightly scattered on the basal membranes of glomeruli (arrows) (electron micrograph). Bar = 1 μm.

Discussion

According to two previous studies (Couchman and Woods 1993; Couchman et al. 1994), the basement membranes of renal glomeruli, Bowman's capsules, and urinary tubule epithelia in mammals contain a substantial amount of heparan sulfate. It is known that sulfated glycoconjugates exhibit positive S-HID reactions in tissues (Hirabayashi 1992) and that heparitinase can specifically degrade heparan sulfates (Hovingh and Linker 1970). In view of these facts and the present results, we suggest that the S-HID reactions of all the basement membranes examined can largely be attributed to heparan sulfate.

In the present study, the positive S-HID reactions of the mesangial matrices, the surface coat of the podocytes, and vascular endothelial cells were not affected by the digestion with heparitinase. According to previous reports (Couchman and Woods 1993; Couchman et al. 1994), the mesangial matrices contain both heparan sulfate and isomeric chondroitin sulfates. Laitinen et al. (1989) and Kogaya and Nanci (1992) suggested that the cell surface coat of glomerular podocytes contains both sialoglycoproteins and sulfated glycoconjugates. It also has been reported that thrombomodulin bearing a single chondroitin/dermatan sulfate chain is present in the surface coat of vascular endothelial cells (Kjellén and Lindahl 1991; Couchman and Woods 1993). In view of these reports, it is likely that the heparitinase-resistant S-HID-positive reactions in glomeruli are attributable to sulfated glycoconjugates other than heparan sulfate.

This newly developed and relatively simple approach allows precise correlation of light and electron microscopic S-HID reaction products. Such an approach could contribute generally to a precise analysis of acidic glycoconjugates in light and electron microscopic histo- and cytochemistry.

References

Couchman

Beavan

McCarthy

(1994) Glomerular matrix: synthesis, turnover and role in mesangial expansion. Kidney Int 45:328–335.

Couchman

Woods

(1993) Structure and biology of pericellular proteoglycans. In Roverts

Mecham

, eds. Cell Surface and Extracellular Glycoconjugates. Structure and Function. New York, Academic Press, 33–82.

Hirabayashi

(1992) Light-microscopic detection of acidic glycoconjugates with sensitized diamine procedures. Histochem J 24:409–418.

Hovingh

Linker

(1970) The enzymatic degradation of heparin and heparitin sulfate. III. Purification of a heparitinase and a heparinase from Flavobactera. J Biol Chem 245:6170–6175.

Kjellén

Lindahl

(1991) Proteoglycans: structures and interactions. Annu Rev Biochem 60:443–475.

Kogaya

Nanci

(1992) Post-embedding staining with high-iron diamine-thiocarbohydrazide-silver proteinate and its application to visualizing sulfated glycoconjugates in cryofixed kidney and cartilage. J Histochem Cytochem 40:1257–1267.

Kushida

(1981) A new method for both light and electron microscopy of identical sites in semi-thin tissue sections embedded in GMA, Quetol 523 and methyl methacrylate. J Electron Microsc 30:77–80.

Kushida

Nagato

Iijima

Kushida

(1993) Correlative light and electron microscopy of the same sections embedded in HPMA, Quetol 523 and MMA. Okajimas Folia Anat Jpn 69:277–288.

Laitinen

Lehtonen

Virtanen

(1989) Differential expression of galactose and N-acetylgalactosamine residues during fetal development and postnatal maturation of rat glomeruli as revealed with lectin conjugates. Anat Rec 223:311–321.

10.

Nagato

Kushida

(1985) Application of ruthenium red staining to the histochemical demonstration of mast cells in semi-thin sections embedded in glycol methacrylate, Quetol 523 and methyl methacrylate. J Electron Microsc 34:139–143.

11.

Nagato

Kushida

(1984) A method for histochemical localization of glycosaminoglycans in semi-thin tissue sections embedded in GMA, Quetol 532 and methyl methacrylate. J Electron Microsc 33:252–254.

12.

Rieder

Bowser

(1983) Factors which influence light microscopic visualization of biological material in sections prepared for electron microscopy. J Microsc 132:71–80.

13.

Rieder

Bowser

(1985) Correlative immunofluorescence and electron microscopy on the same section of Epon-embedded material. J Histochem Cyotochem 33:165–171.